Tool-hook integration
Agno
Wrap an Agno tool with a FreshCtx protected-action boundary using agno_tool_hook() or agno_async_tool_hook().
Agno integration details →Integrations
Where FreshCtx sits in your agent runtime
FreshCtx keeps the freshness boundary separate from the agent framework. The framework changes; the invariant does not: declared evidence is revalidated immediately before the protected action.
Action-node integration
LangGraph
Map langgraph_action_node() around the graph node that performs the external write, keeping dependency IDs in graph state.
LangGraph integration details →Tool input guardrail
OpenAI Agents SDK
Use openai_agents_tool_guardrail() as a tool input guardrail on a custom SDK function tool.
OpenAI Agents SDK integration details →before_tool_callback integration
Google Agent Development Kit
Use google_adk_tool_callback() as an ADK agent's before_tool_callback to skip a protected tool body when evidence is stale.
Google Agent Development Kit integration details →tools/call execution boundary
MCP Guard
An opt-in extension for the official MCP Python SDK v2 that revalidates evidence at the native tools/call boundary.
MCP Guard integration details →Third-party names and logos are trademarks of their respective owners. Their appearance identifies compatible integration surfaces and does not imply endorsement or partnership.